Re: Disadvantage of Non-parametric vs. Parametric Test
Alex Yu wrote: Disadvantages of non-parametric tests: Losing precision: Edgington (1995) asserted that when more precise measurements are available, it is unwise to degrade the precision by transforming the measurements into ranked data. So this is an argument against rank-based nonparametric tests rather than nonparametric tests in general. In fact, I think you'll find Edgington highly supportive of randomization procedures, which are nonparametric. In fact, surprising as it may seem, a lot of the location information in a two sample problem is in the ranks. Where you really start to lose information is in ignoring ordering when it is present. Low power: Generally speaking, the statistical power of non-parametric tests are lower than that of their parametric counterpart except on a few occasions (Hodges Lehmann, 1956; Tanizaki, 1997). When the parametric assumptions hold, yes. e.g. if you assume normality and the data really *are* normal. When the parametric assumptions are violated, it isn't hard to beat the standard parametric techniques. However, frequently that loss is remarkably small when the parametric assumption holds exactly. In cases where they both do badly, the parametric may outperform the nonparametric by a more substantial margin (that is, when you should use something else anyway - for example, a t-test outperforms a WMW when the distributions are uniform). Inaccuracy in multiple violations: Non-parametric tests tend to produce biased results when multiple assumptions are violated (Glass, 1996; Zimmerman, 1998). Sometimes you only need one violation: Some nonparametric procedures are even more badly affected by some forms of non-independence than their parametric equivalents. Testing distributions only: Further, non-parametric tests are criticized for being incapable of answering the focused question. For example, the WMW procedure tests whether the two distributions are different in some way but does not show how they differ in mean, variance, or shape. Based on this limitation, Johnson (1995) preferred robust procedures and data transformation to non-parametric tests. But since WMW is completely insensitive to a change in spread without a change in location, if either were possible, a rejection would imply that there was indeed a location difference of some kind. This objection strikes me as strange indeed. Does Johnson not understand what WMW is doing? Why on earth does he think that a t-test suffers any less from these problems than WMW? Similarly, a change in shape sufficient to get a rejection of a WMW test would imply a change in location (in the sense that the "middle" had moved, though the term 'location' becomes somewhat harder to pin down precisely in this case). e.g. (use a monospaced font to see this): :. .: ::. = .:: ... ... a b a b would imply a different 'location' in some sense, which WMW will pick up. I don't understand the problem - a t-test will also reject in this case; it suffers from this drawback as well (i.e. they are *both* tests that are sensitive to location differences, insensitive to spread differences without a corresponding location change, and both pick up a shape change that moves the "middle" of the data). However, if such a change in shape were anticipated, simply testing for a location difference (whether by t-test or not) would be silly. Nonparametric (notably rank-based) tests do have some problems, but making progress on understanding just what they are is difficult when such seemingly spurious objections are thrown in. His preference for robust procedures makes some sense, but the preference for (presumably monotonic) transformation I would see as an argument for a rank-based procedure. e.g. lets say we are in a two-sample situation, and we decide to use a t-test after taking logs, because the data are then reasonably normal... in that situation, the WMW procedure gives the same p-value as for the untransformed data. However, let's assume that the log-transform wasn't quite right... maybe not strong enough. When you finally find the "right" transformation to normality, there you finally get an extra 5% (roughly) efficiency over the WMW you started with. Except of course, you never know you have the right transformation - and if the distribution the data are from are still skewed/heavy-tailed after transformation (maybe they were log-gamma to begin with or something), then you still may be better off using WMW. Do you have a full reference for Johnson? I'd like to read what the reference actually says. Glen
UNSUBSCRIBE UNSUBSCRIBE
\\|// (o o) +--oOO--(_)--OOo-+ | Jose Manuel Rojo | tlf : +++ 91 + 5-64-29-63| | C.S.I.C , SPAIN | Email: [EMAIL PROTECTED]| ++ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-==-=-=-==-=-=-=- o \ o / _o __|\ / |__ o _ \ o / o /|\| /\ __\o\o|o/ o/__ /\ | /|\ / \ / \ | \ /) |( \ /o\ / ) | (\ / |/ \ / \ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Re: Scale Reliability
Brett - Although you may have a conceptual reason for summing the four item scores to form a composite, the low level of internal consistency that these items display suggests that they may not be assessing a unitary construct. So, to sum these four items and label this score as a single variable or construct (locus of control?) may be problematic. Have you assessed the dimensionality of these items? If you find that these items do not strongly load on a single, unitary factor, you may want to consider breaking the items up, rather than summing them. John
Re: Scale Reliability
The fact that the shorter scale has low internal consistency doesn't necessarily mean that the 4 items in question are not unidimensional. It may just be that the measurement error is large relative to their covariance. Given that the four items in question are drawn from a scale with established internal consistency, I'd suspect they probably are measuring the same thing - only not measuring it very well. purnima. To: cc: From: John Donovan [EMAIL PROTECTED]/ Date: 12/07/99 01:09:40 PM Subject: Re: Scale Reliability Brett - Although you may have a conceptual reason for summing the four item scores to form a composite, the low level of internal consistency that these items display suggests that they may not be assessing a unitary construct. So, to sum these four items and label this score as a single variable or construct (locus of control?) may be problematic. Have you assessed the dimensionality of these items? If you find that these items do not strongly load on a single, unitary factor, you may want to consider breaking the items up, rather than summing them. John
Re: stats packages for Unix
This is great ... I had lost my reference to this site and product. Thanks! On Sat, 4 Dec 1999, KO wrote: Hello: I would suggest R (aka GNU-S). You cane find more information about it at http://www.stat.cmu.edu/R/CRAN/ I highly recommend it. Sincerely, Kouros Owzar ([EMAIL PROTECTED]) Bob Hayden [EMAIL PROTECTED] wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... Any advice on basic stats. packages for Unix/Linux? We used Minitab on a DEC alpha box using DEC Unix. That was fine by me but the DEC was too expensive to maintain. So, we replaced it with a FreeBSD system runing on a pile of Intel boxes. But, Minitab does not run on that, so we started using the PC and iMac versions. However, the iMac lab is a disaster (local problem, not a Mac problem, I think) and we hate supporting three platforms. The software must be VERY easy to use and not cost much more than Minitab, which is quite cheap for a site license. We have a few hundred students per year using the software. Most of them have pretty limited math., computer and study skills. _ | | Robert W. Hayden | | Department of Mathematics / | Plymouth State College MSC#29 | | Plymouth, New Hampshire 03264 USA | * | Rural Route 1, Box 10 /| Ashland, NH 03217-9702 | ) (603) 968-9914 (home) L_/ [EMAIL PROTECTED] fax (603) 535-2943 (work)
Re: Software for logistic regression
SYSTAT has an excellent log regression procedure and it may be cheap enough for you. On Sun, 5 Dec 1999, FourCubed wrote: Hello everyone! Can anyone tell me the least expensive software for performing a (conditional) logistic regression. SPSS and SAS are too expensive, and I don't have access to a university computer with existing software. Thank you. Steve [EMAIL PROTECTED]
Re: Help for my dissertation
First, check to see if your school doesn't have a stat lab available. Most major universities have a site where graduate students and others involved in empirical research can get free assistance. Often, it is in the form of anothe, but more advanced, graduate student, but many of these folks are perfectly capable of looking over your proposal and can give helpful hints. Second, you might want to consider adding a statistically knowledgeable person to your committee if it is not too late. If I were you, I would get the statistical guidance BEFORE you start collecting data or whatever methodology you envision. Lastly, you can go to the statistics department at your school and get the names of consultants available for looking over your materials. Good Luck j. williams In article 82j1a2$6s3$[EMAIL PROTECTED], [EMAIL PROTECTED] wrote: I am a 3rd year student carrying out resaerch for my dissertation. I require help with stats - primarily to confirm that I amusing the correct methods to test my hypothesis and secondly to understand them once I have collected my data. I am prepared to pay for some ones time if they could help me. Is there anyone out there that could help? Sent via Deja.com http://www.deja.com/ Before you buy.
STATISTICA group
Dear STATISTICA users, in order to subscribe to the mailing list/group for STATISTICA users, visit http://www.computer.privateweb.at/Statistica/ and type just in your e-mail adress and follow the instructions. Then you will be added to our list. My very best wishes Dr. Wolfgang Hitzl (group moderator)
SDT: What is D-sub-A and Sakitt's D?
[Please forgive me if this esoteric question was answered before in this newsgroup] I am using Systat 9.0 for my master's thesis data--the nature of my analyses depend heavily on Signal Detection Theory. Therefore, of course I am using the Signal Detection Analysis program in Systat. It seems to me that using d' is not as good as D-sub-a or Sakitt's D -- the subjects' responses I am analyzing are confidence ratings, on a scale from one-to-eight -- as there tends to be greater fluctuations/variability of d-prime as opposed to the other measures. In this way, I'm committed to D-sub-a thinking it is more robust and consistent. But I need to know how D-sub-A and Sakitt's D differ from d'. I haven't been able to find information on this from any other source thus far. Can anyone tell me what these measures are and how they differ from d'? Thanks, V.
Re: Help for my dissertation
I'd second the comments of finding local help; that's definitely the best option. If, for whatever reason, that's not available, then, and only then, consider looking for someone at a distance to help out. If you do have to do that, offer more detail about what kind of study you're doing: the type of data, how it's collected, how it's analyzed, what questions you want to investigate, etc.; that's all necessary before anyone (remote or local) can determine what you need and whether they can help you. Steve Albert
